According to MSCI, 95% of surveyed firms expect to increase their AI investment over the next three years. However, only 27% believe that wealth management is leading other financial-services sectors in AI adoption. EY reports a similar gap: AI is already being used across multiple areas of wealth and asset management, but far fewer firms are seeing a substantial impact on their business.

In other words, companies are adopting AI much faster than they are learning how to get real value from it.

So far, the clearest benefits have come from improving everyday work rather than replacing wealth managers or handing investment decisions over to AI. Current uses include meeting preparation, documentation, client follow-ups, information analysis, compliance and other administrative tasks. EY describes the next stage as a move from “superficial adoption to deep integration” — making AI part of the systems and processes that advisers already use every day.

But this exposes a more fundamental problem.

Wealth-management firms already rely on a complex mix of CRM systems, portfolio and reporting platforms, financial-planning tools and other specialised software. Client information is often spread across these systems, historical data may be inconsistent, and information still has to be transferred or reconciled manually. MSCI identifies fragmented data as one of the main obstacles to wider AI adoption. Capgemini points to the same problem: many wealth-management firms still do not have a single, complete view of their clients.

Simply adding another AI tool is unlikely to solve this.

Instead, a different technology model is beginning to appear. Existing platforms can continue to store data and perform their specialised functions, while AI sits above them — bringing information together, analysing structured and unstructured data and helping people work across different systems. Capgemini calls this an “intelligence layer.”

We can already see this happening.

In September 2026, Anthropic launched Claude for Financial Advisors. Instead of building another wealth-management platform, Anthropic connected Claude to systems that advisers already use, including Charles Schwab, Addepar, Envestnet, iCapital, Orion and Wealthbox. Claude can use information from these systems to help prepare for client meetings, review portfolios, conduct research, prepare documents and handle follow-up work.

This may be an early sign of a broader change in wealth-management technology: AI is moving from being a separate tool towards becoming a layer connecting existing software, data and workflows.

So what will the next generation of wealth-management technology look like: new AI-native platforms, or an intelligent AI layer connecting the specialised systems wealth managers already use?

CGL Group is currently exploring this question as part of its research into AI in wealth management and family offices, focusing on emerging technologies, software solutions and their practical applications.